AI Safety Engineering is a six-week specialisation for the technical practitioners who build, deploy and assure AI systems where failure has real-world consequences. It moves past principles into engineering: how AI systems fail, how to design controls that catch failure before it reaches a person, how to test a system adversarially, how to suppress harmful outputs, and how to monitor a live system so safety is a measured property rather than a hope. By the end you will be able to reason about failure modes, build a safety control stack, run a red-team, and stand up a safety assurance programme that an engineer, an auditor and the NDPC could all trust.
For: ML engineers, data scientists, MLOps and platform engineers, technical risk and assurance leads, and AI product owners deploying safety-critical AI in Nigerian banking, health, telecoms and public-sector contexts.
This is a Specialisation-level course. It assumes you can read code and reason about models; it teaches you to make them safe under the NDPA 2023, NITDA guidance, and international practice such as the NIST AI Risk Management Framework and ISO/IEC 42001.